Pregnancy does not reduce the risk of recurrence among women who have previously experienced depressive illness and the advent of new episodes during pregnancy raises particular problems. Concerns over the possible teratogenicity of medications in general leave many women reluctant to continue preexisting antidepressant prophylaxis or to accept new trials of conventional antidepressant treatment and there is accumulating evidence that the SSRIs have short-term adverse effects on the newborn. The antidepressant effects of omega-3 polyunsaturated fatty acid (PUFA) supplementation may offer a particularly appropriate alternative to conventional therapy for depressive episodes that occur during pregnancy. The nutritional needs of the fetus increase the likelihood of omega-3 PUFA deficits in the mother but access to adequate omega-3 PUFAs but fish intake is limited due to concerns over mercury levels. Antidepressant trials of omega-3 supplementation that have described significant benefits over placebo include one that targeted pregnant women and yielded a large effect size. Other trials, however, have failed to show clear antidepressant effects and meta-analyses have yielded no explanations for these inconsistencies. A clear possibility is that the studies with positive results involved subjects who more likely to benefit from omega-3 supplementation but the characteristics of such individuals are entirely unknown. Numerous case-control studies have associated depressive illness with lower tissue concentration of omega-3 PUFAs and with higher ratios of omega-6 to omega-3. Such measures may well identify individuals likely show antidepressant effects from supplementation. The likelihood that omega-3 PUFAs exert antidepressant effects via modulation of the inflammatory cascade, and the extensive evidence that high levels of cytokines characterize individuals with depressive disorders, indicate that these measures too may help to select those most likely to benefit from treatment with omega-3 PUFAs. The above considerations lead us to propose a developmental (R21) grant with two goals involving two overlapping subject groups. The larger would consist of 120 women in their first trimester of pregnancy who are not currently in major depressive episodes but who are at high risk of recurrence; they would be used to identify measures of PUFA tissue concentration and inflammatory marker status as risk factors for later onsets of major depressive episodes. A second group would consist of 60 women who begin pregnancy in depressive episodes or who develop episodes in their first two trimesters but who choose not to take conventional antidepressant therapy; they would be used to test PUFA tissue concentration and inflammatory measures as predictors of response to omega-3 supplementation mono- therapy. Together the results allow for the personalization both of acute antidepressant treatment and of omega-3 prophylaxis for women at high risk for perinatal depressive disorder and would provide important information for the design of larger controlled trials of omega-3 supplementation.